The White Sox bullpen has been a frequent subject of discussion this year, mostly out of a sense that they haven't lived up to expectations. Granted, a group that came into the year with the goal of blowing zero leads beyond the 6th inning was sure to disappoint some segment of the fanbase. However, there has been a real dip in performance with the group as a whole since the second half of the season.
To be clear, the bullpen hasn't been bad in the second half. The group is ranked sixth in the AL in WAR since the All-Star break and ranks fourth in MLB for the entire season, but the ordinary performances they have provided recently do not square with their talent on paper. Expectations were heightened after the team acquired Ryan Tepera and Craig Kimbrel from the Cubs, and individual performances from every major contributor out of the pen have been scrutinized in the last month and a half. Since three members have their own nickname and t-shirt design, I'm going to see if there's anything revealing in the recent struggles (real or exaggerated) of Michael Kopech, Craig Kimbrel and Liam Hendriks.
Michael Kopech
In a general sense, there have been two different Kopechs this year: pre-injury version, and the version that has struggled since late July. The bulk of this section will be comparing these two different stretches for him.
First, Kopech's fastball does not appear to be a major issue. While it has been hit harder than it did earlier in the season, the xwOBA on the pitch is actually lower than the first two months of the year, at .196 versus .260. This has happened alongside a 100 point jump in his actual wOBA, suggesting some bad luck at play here. He is also locating the pitch in attack zones similarly now to how he has all year, and throwing the pitch nearly 2 MPH harder since late July than before his injury. Moreover, nothing with his fastball suggests there is a major issue that will impact his performance.
Kopech's problem in this recent stretch that could cause him more trouble involves his slider. He's leaving the pitch in the heart of the zone 32.4% of the time in his recent stretch after throwing it there 21.8% the first two months of the year. Even at his best, Kopech has not succeeded when hanging the slider over the plate. The pitch doesn't get swings and misses there, with only 7 whiffs in the heart of the zone all year. Called strikes were the most frequent outcome the first two months, so he could survive with the pitch there occasionally. But hitters are getting the bat to the ball two thirds of the time the pitch is thrown now, and doing damage when it's put in play.
Location isn't the only thing different about the pitch since coming back from injury, as the pitch also has a different shape. The spin axis on his slider is more flat, with an average of 56.8 degrees since late July compared to 67.5 degrees beforehand. This axis tilt has given the pitch more vertical drop, causing it to have negative vertical movement (additional drop beyond the effects of gravity) 48% of the time compared to 38% before his injury.
This additional movement seems to have brought some positive effects. The pitch has generated more whiffs since Kopech's return, which has in turn contributed to a greater strike/ball ratio. However, the frequency he's leaving it up in the zone might indicate that he's having trouble locating his slider with its new movement. The question about this development is whether it is intentional, or the byproduct of something else going on with his delivery.
Since I'm not sure how often pitch design is altered mid-season, I wouldn't rule out the possibility that a mechanical change is leading him to unintentionally impart a different type of spin on the ball. Although it's harder to see without a side view, it's clear that there is a difference in his lower half since returning from injury. He is sitting more into his quads now than he did previously, which is causing his center of mass to be shifted back and towards the first base side. We can see the difference comparing a pitch from April on the left, and August on the right.
His knee appears in a position where it is going to buckle down rather than drive forward, and the instability in his weight distribution looks like a step towards the old version of Reynaldo López rather than towards the other pitchers on the staff.
The weight distribution in his new form can be seen based on where his back foot is as well. In this first image from April, when he comes close to planting his front foot and firing his hips, his back foot is sliding over its side, showing that his momentum is already taking his center of mass towards the plate and bringing the back side with.
However, looking at the same point in his delivery more recently, the balls of his feet are still planted into the ground, showing that he has more weight distributed back as his hips are ready to open.
Knowing that Kopech's center of mass seems to be shifted back and lower than before, it's fair to wonder if it has taken his release point in this direction. Sure enough, after adjusting for pitch type and batter handedness, there is a slight movement towards the first base side, and a more noteworthy drift downward. However, listing out average values for his release points wouldn't tell the full story. Rather than a clear drop late in the season, the vertical release point has become increasingly volatile as the season has progressed.
He is generally repeating his delivery less consistently than early on, but it's harder to explain why. Fatigue would be the obvious answer, but it's difficult to argue that when he's throwing his fastball harder now than earlier in the year, and doing better job of maintaining it within outings. It could be that Kopech has had to make lower half adjustments to avoid reaggravating the hamstring injury that sidelined him for a month, but this doesn't seem like it's produced a better delivery for him long term.
At the end of the day, we know his delivery is slightly different, his slider has a different shape to it, and he's leaving sliders in the zone more, while it's difficult to say how much of this is intentional. The August box scores would tell you he should go back to what he did originally, but there's enough positives mixed in with his recent outings that it's hard to pinpoint any individual action as something making him a worse pitcher.
Craig Kimbrel
For Kimbrel, I suggest you check out this Reddit post first, highlighting some of the areas Kimbrel has faltered in since coming to the Sox. It mainly talks about him throwing fewer strikes and his curveball location not being as precise as before, but his fastball location has been lacking as well. As you can see below, he is drifting away from the corners of the zone towards the middle.
This may explain the majority of his struggles since joining the White Sox, as the biggest difference since leaving the Cubs is a higher swing percentage on his pitches in the zone, and a lower swing percentage on pitches out of the zone. Opponents are swinging about 9% more often in the zone and 5% less out of the zone. With his Whiff% identical between the two teams, the swing rate changes are leading to more quality contact against him. Hitters seem to be recognizing his pitches better, and throwing the fastball closer to the middle of the zone and curveball farther away appear to be reasonable explanations.
Kimbrel has also talked about how he is not getting 'behind' the ball as much in recent weeks, saying this has been a fault of his on bad outings. You can listen for yourself, but the way I read these comments is that he thinks his delivery sometimes becomes overly rotational, and he is working to maintain a stronger linear component. While I'm glad he seems to know what's ailing him, it's hard to find anything in the video or data directly supporting this. The only part of his delivery that's clearly different is that his release extension on fastballs is lower on the Sox compared to the Cubs this year. However, his current release extension numbers are more in line with career norms, and his first half with the Cubs this year sticks out as more of an anomaly.
It's more likely Kimbrel is working with more (and better) information on his struggles, and his recent performances suggest that he's turning a corner.
Since Kimbrel's peripherals are largely the same with both the Sox and Cubs this year, the one adjustment I was going to suggest was for Kimbrel to start off more at bats with a curveball. He threw each pitch about half the time on first pitch with the Cubs, but recorded a run value per 100 of -3.67 with the pitch on 0-0 counts, versus .5 with the fastball. He raised that fastball rate to about 70% since joining the Sox with worse results, but it appears he's already begun making a change. I gathered this data before his appearance on August 31st, but since then he has started off 5 out of 7 hitters he's faced with a curveball.
Liam Hendriks
This has been a very good year from Liam Hendriks, and that doesn't feel as controversial to say as it once did. There was a small wave of panic after his appearances against the Yankees, but his run of good outings since then and recognition of a pitch tipping issue have calmed many of these fears. Even more so than Kimbrel, Liam's peripherals are very similar to his best years. Not only do the K/BB rates match, but he is also locating like he always has (although locating may be a strong word).
I imagine he has an approach to individual hitters, but there isn't an overarching precision to his game. These heatmaps show a guy who simply throws his heat up in the zone and challenges hitters to hit it, which works quite often. He does hit the corners with his slider, but that pitch doesn't hurt him to leave in the zone either. His Whiff% has climbed north of 30% with the pitch in the zone this year, and his xwOBA there is below .200.
Basically, Hendriks has been the same guy he's been his whole career in terms of approach, and the results have largely followed. The only difference is hitters have done some damage on balls he's left in the zone, though this hasn't translated towards more base hits aside from the long ball. Until that happens, it isn't worth being overly concerned about him, and we can enjoy counting how many more Home Run Derby participants he can give up homers to.
(Photo by Dan Hamilton/USA TODAY Sports)
